Attacking Spectrum Sensing With Adversarial Deep Learning in Cognitive Radio-Enabled Internet of Things

Cognitive radio-based Internet of Things (CR-IoT) network provides a solution for IoT devices to efficiently utilize spectrum resources. Spectrum sensing is critical problem in CR-IoT network, which has been investigated extensively based on deep learning (DL). Despite the unique advantages DL sensing, black-box and unexplained properties neural networks may lead many security risks. This article considers fusion traditional interference methods data poisoning an attack method training machine tool. We propose new adversarial reducing accuracy DL-based systems. introduce novel design jamming waveform whose capability reinforced by poisoning. Simulation results show that significant performance enhancement higher mobility can be achieved compared with white-box methods.